Occupancy sensor

Abstract

System and techniques for an occupancy sensor are described herein. Images from a camera can be received. Here, the camera has a certain a field of view. A proximity indicator from a proximity detector can be received when an object enters the field of view. The images from the camera are processed to provide an occupancy indication. A first technique is used for the occupancy indication in the absence of the proximity indicator and a second technique is used otherwise.

What is claimed is: 
     
       1. A device, comprising:
 a camera with a field of view; 
 a proximity detector arranged to detect an object in the field of view, the proximity detector being separate from the camera, and a sensing area of the proximity detector overlapping with the field of view of the camera and at least a portion of the field of view of the camera is non-overlapping with the sensing area; and 
 a controller to:
 receive images from the camera; 
 receive a proximity indicator from the sensing area of the proximity detector overlapping with the field of view of the camera, wherein the proximity indicator originates from the proximity detector and indicates absence or presence of the object within the sensing area that overlaps with the field of view of the camera, wherein the proximity detector has a threshold distance and provides the proximity indicator in response to detecting the object within the threshold distance; and 
 process the images from the camera to provide an occupancy indication based on the proximity indicator, wherein a first occupancy detection technique is used in the absence of the object from the sensing area, and a second occupancy detection technique is used when the object is present in the sensing area, wherein the first technique is different from the second technique and uses lower power than the second technique. 
 
 
     
     
       2. The device of  claim 1 , wherein the threshold distance is a fixed threshold distance. 
     
     
       3. The device of  claim 2 , wherein the proximity detector is an active infrared (IR) proximity detector. 
     
     
       4. The device of  claim 1 , wherein the images are two-dimensional images. 
     
     
       5. The device of  claim 1 , wherein the first technique includes the controller to:
 identify the object as a foreground object using the images from the camera; and 
 provide the occupancy indication when the foreground object is beyond a threshold size. 
 
     
     
       6. The device of  claim 5 , wherein to identify the foreground object the controller is to apply a mixture of Gaussians to a sequence of the images. 
     
     
       7. The device of  claim 5 , wherein the second technique includes the controller to:
 apply pattern matching of a classifier to the foreground object when the foreground object is beyond the threshold size; and 
 provide the occupancy indication based on the result of the classifier. 
 
     
     
       8. The device of  claim 7 , wherein the classifier is trained to identify an exclusion object and wherein the occupancy indication is provided when the classifier does not identify the foreground object. 
     
     
       9. The device of  claim 8 , wherein an exclusion object includes a flying animal. 
     
     
       10. The device of  claim 7 , wherein the classifier is trained to identify an inclusion object and wherein the occupancy indication is provided when the classifier does identify the foreground object. 
     
     
       11. A non-transitory machine-readable medium including instructions that, when executed by a machine, cause the machine to perform operations comprising:
 receiving images from a camera, the camera having a field of view; 
 receiving a proximity indicator from a sensing area of a proximity detector overlapping with the field of view of the camera, the proximity detector arranged to detect an object in the field of view, wherein the proximity detector is separate from the camera and wherein the proximity indicator indicates absence or presence of the object within the sensing area overlapping with the field of view of the camera and at least a portion of the field of view of the camera is non-overlapping with the sensing area, wherein the proximity detector has a threshold distance and provides the proximity indicator in response to detecting the object within the threshold distance; and 
 processing the images from the camera to provide an occupancy indication based on the proximity indicator, wherein a first occupancy detection technique is used in the absence of the object from the sensing area, and a second occupancy detection technique is used when the object is present in the sensing area, wherein the first occupancy detection technique is different from the second occupancy detection technique and uses lower power than the second technique. 
 
     
     
       12. The non-transitory machine-readable medium of  claim 11 , wherein the threshold distance is a fixed threshold distance. 
     
     
       13. The non-transitory machine-readable medium of  claim 12 , wherein the proximity detector is an active infrared (IR) proximity detector. 
     
     
       14. The non-transitory machine-readable medium of  claim 11 , wherein the images are two-dimensional images. 
     
     
       15. The non-transitory machine-readable medium of  claim 11 , wherein the first technique includes:
 identifying the object as a foreground object; and 
 providing the occupancy indication when the foreground object is beyond a threshold size. 
 
     
     
       16. The non-transitory machine-readable medium of  claim 15 , wherein identifying the foreground object includes applying a mixture of Gaussians to a sequence of the images. 
     
     
       17. The non-transitory machine-readable medium of  claim 15 , wherein the second technique includes:
 applying pattern matching of a classifier to the foreground object when the foreground object is beyond the threshold size; and 
 providing the occupancy indication based on the result of the classifier. 
 
     
     
       18. The non-transitory machine-readable medium of  claim 17 , wherein the classifier is trained to identify an exclusion object and wherein the occupancy indication is provided when the classifier does not identify the foreground object. 
     
     
       19. The non-transitory machine-readable medium of  claim 18 , wherein an exclusion object includes a flying animal. 
     
     
       20. The non-transitory machine-readable medium of  claim 17 , wherein the classifier is trained to identify an inclusion object and wherein the occupancy indication is provided when the classifier does identify the foreground object. 
     
     
       21. A method for providing occupancy indication, the method comprising:
 by a computing device comprising a proximity detector: 
 receiving images from a camera, the camera having a field of view; 
 receiving a proximity indicator from a sensing area of the proximity detector overlapping with the field of view of the camera, the proximity detector arranged to detect an object in the field of view, wherein the proximity detector is separate from the camera and wherein the proximity indicator indicates absence or presence of the object within the sensing area overlapping with the field of view of the camera and at least a portion of the field of view of the camera is non-overlapping with the sensing area, wherein the proximity detector has a threshold distance and provides the proximity indicator in response to detecting the object within the threshold distance; and 
 processing the images from the camera to provide an occupancy indication based on the proximity indicator, wherein a first occupancy detection technique is used in the absence of the object from the sensing area, and a second occupancy detection technique is used when the object is present in the sensing area, wherein the first occupancy detection technique is different from the second occupancy detection technique and uses lower power than the second occupancy detection technique. 
 
     
     
       22. The method of  claim 21 , wherein the threshold distance is a fixed threshold distance. 
     
     
       23. The method of  claim 21 , wherein the proximity detector is an active infrared (IR) proximity detector. 
     
     
       24. The method of  claim 21 , further comprising:
 identifying the object as a foreground object; and 
 providing the occupancy indication when the foreground object is beyond a threshold size. 
 
     
     
       25. The method of  claim 24 , wherein identifying the foreground object includes applying a mixture of Gaussians to a sequence of the images.